Simulated Annealing Algorithm To Solve The Static Rwa Problem

In optical network based on WDM, traffic engineering (TE) is mapping traffic flows on the physical topology of optical network according to user's requirement, in order to optimize the network utility, avoid jamming and improve the network reliability. Accompany with the continuous development of OXC (Optical Cross Connect) and OADM (Optical Add/Drop Multiplexer), optical network based on DWDM (Dense Wavelength Division Multiplexing) is no longer restricted in the peer-to-peer transmission, but it spans to the area of network interconnecting. Thus, the network resource allocation based on need and optimization is desperate. Resource allocation based on need is to allocate limited network resource according to the need of different users, in order to sate various applications. The aim of optimization is to allocate resource more reasonably, in order to minimize the total cost of the whole net, to maximize the utility. To solve these two problems mentioned above and to sate the need of traffic QoS, TE is needed in the WDM optical network.It is defined as RWA problem that how to rout optical path and allocate net resource according to a set of request in WDM network. It is the key problem that how to switch an effective algorithm and available protocol to make up light-paths as well as solving RWA problem. That is very important to study the net resource allocation , network management and control. We can divide RWA into static and dynamic according to different traffic request. A lot of method have put to met the demand such as linear-programing or some heuristic algorithms.This paper describes solving the static RWA problem, an aspect of TE problem in optical network, with simulated annealing algorithm. The static RWA problem is to allocate proper route and wavelength for a set of static requests on a WDM network with and without OXCs which physical topology is known. To compare the results of simulated anneal algorithm on the static RWA problem, a heuristic algorithm is introduced as reference. In this paper, a network model based on real network situation is studied. Then the result of certain k-shortest path algorithm is introduced as a set to be chosen by both the simulated anneal algorithm and heuristic algorithm, concerning the end-to-end traffic requirement and restriction on link bandwidth. Afterwards, the effect of the value k on the outcome of the simulated anneal algorithm is researched. The network resource is allocated, while the study of link protection problem on traffic routing is introduced and two types of protection, shared protection and dedicated...